Limit of detection and threshold for positivity of the Centers for Disease Control and Prevention assay for factor VIII inhibitors

Abstract

Background The Bethesda assay for measurement of factor VIII inhibitors called for quantification of positive inhibitors by using dilutions producing 25-75% residual activity (RA), corresponding to 0.4-2.0 Bethesda units, with the use of 'more sensitive methods' for samples with RA closer to 100% being recommended. The Nijmegen modification (Nijmegen-Bethesda assay [NBA]) changed the reagents used but not these calculations. Some specimens negative by the NBA have been shown to have FVIII antibodies detectable with sensitive immunologic methods. Objective To examine the performance at very low inhibitor titers of the Centers for Disease Control and Prevention (CDC)-modified NBA (CDC-NBA), which includes preanalytic heat inactivation to liberate bound anti-FVIII antibodies. Methods Specimens with known inhibitors were tested with the CDC-NBA. IgG4 anti-FVIII antibodies were measured by fluorescence immunoassay (FLI). Results Diluted inhibitors showed linearity below 0.4 Nijmegen-Bethesda units (NBU). With four statistical methods, the limit of detection of the CDC-NBA was determined to be 0.2 NBU. IgG4 anti-FVIII antibodies, which correlate most strongly with functional inhibitors, were present at rates above the background rate of healthy controls in specimens with titers ≥ 0.2 NBU and showed an increase in frequency from 14.3% at 0.4 NBU to 67% at the established threshold for positivity of 0.5 NBU. Conclusions The CDC-NBA can detect inhibitors down to 0.2 NBU. The FLI, which is more sensitive, demonstrates anti-FVIII IgG4 in some patients with negative (< 0.5) NBU. The sharp increase in IgG4 frequency between 0.4 and 0.5 NBU validates the established threshold for positivity of ≥ 0.5 NBU for the CDC-NBA, supporting the need for method-specific thresholds.